The complexity of eukaryotic cells requires intracellular organization, coordination, and locomotion. To overcome these challenges, cells utilize ATP-driven molecular motors, which transport intracellular components unidirectionally along cytoskeletal tracks. Kinesin and cytoplasmic dynein motors facilitate bidirectional transport of a variety of cargos by moving towards the plus- and minus-ends of microtubules (MTs), respectively. Detailed mechanistic models exist for kinesin, but the mechanism and regulation of dynein motility are still emerging. We found that S. cerevisiae dynein walks on a MT through uncoordinated stepping of its two catalytic domains and its mechanism of action differs significantly from the coordinated hand-over- hand stepping of kinesin. Surprisingly, despite recent advances in structural characterization of dynein, the molecular origin of its strong directional preference to move towards the MT minus-end remains unclear. Recently, a recombinant expression system was developed for human dynein, opening the doors for detailed studies of its molecular mechanism for the first time. Surprisingly, human dynein exhibited only short possessive runs and produces significantly lower forces than S. cerevisiae dynein in vitro, inconsistent with the ability of human dynein to transport large intracellular cargos over long distances inside cells. New work has revealed that processivity of human dynein is activated when it forms a 2.5 MDa ternary complex (referred to as DDB) with its cofactor dynactin and a cargo binding adaptor BicD2. In our preliminary work, we showed that dynactin and BicD2 also significantly enhance human dynein's force generation, suggesting that the DDB complex is a strong motor and a formidable opponent of kinesin when attached to the same cargo. The goal of this proposal is to dissect the mechanism of active human dynein complexes and determine how dynactin and BicD2 regulate dynein's ability to compete against kinesin-1 during bidirectional cargo transport. We have three specific aims. First, using protein engineering and single-molecule imaging, we will identify the mechanical components of dynein that give rise to its minus-end directed motility. We will also solve the MT-bound structure of reverse directionality constructs via cryo-electron microscopy (cryoEM) to reveal the structural basis of dynein directionality. Second, we will identify which part(s) of the motor is responsible for its autoinhibition and characterize how dynactin and BicD2 regulate the mechanochemical cycle, stepping pattern and force generation of human dynein. Third, we will reconstitute bidirectional cargo transport on MTs in vitro using purified human kinesin and DDB complexes and reveal the mechanism and regulation of tug-of-war between these motors. Success of our aims will significantly advance the understanding of the fundamental mechanochemistry of human dynein and learn how it achieves retrograde transport of intracellular cargos.